Instructional support system for teaching number sense

Number sense is regarded as a strong predictor of mathematics achievement at formal school. This paper describes the improvement of early math skills using computer-assisted instruction and the effectiveness of an intervention program. An experimental design with control and experimental groups, and pre- and post-intervention measurement was used. Participants were 48 preschool students (aged from 4.91 to 5.91; 21 were males and 27 were female), whose relational and numeracy skills were assessed before and after training. Differences between groups were analyzed. Participants were students from four schools (two public and two private). Schools were in middle-class neighborhood, in 100.000 inhabitants towns. Descriptive analysis, discriminant analysis and hypothesis test were calculated. Null hypothesis of equality between groups was rejected (Wilks' Lambda = .468; X-2 = 31.46; p < .001). Therefore, the difference between the experimental and control groups was statistically significant. Also, discriminant analysis indicated that 83.3% of students were classified correctly in their group. Significant differences between control and experimental group in classifications (p < .001), correspondence (p < .001), counting structured (p < .001) and resulting (p < .001) skills, were found. Results supported the effectiveness of the intervention program, and indicated math skills that were significantly improved by training: classifications and correspondence as relational skills, and counting structured and resulting as numerical skills. Educational implications and future lines of action are discussed.